regarding Honeycutt-Andersen pair index

LAMMPS LAMMPS Mailing List Mirror
1

Dear Axel;

Sorry for the insufficient information. Honeycutt-Andersen method attributes 4 integers considering every two atoms. Each integer defines a property (Number of shared neighbors, number of bonds between those neighbors and so on). Specific pair numbers (e.g. 1551) defines the kind of the pair in a specific manner. It’s like CNA compute but for example there is possible that two different sets of integers belong to FCC structure, for example. Since I have seen lots of such diagrams ( Relative number of a specific Honeycutt-Andersen index Vs. T), I was wondering if there is a way to compute it using Lammps. Thanks again

2

That would require a new peratom compute.
If you want to write it, that would be great.
You could use compute cna/atom or compute
centro/atom as a starting point.

Steve

3

Dear Steve
That’s a great idea. I would let you know if I made any progress.

Regards.

4

Dear Steve
That's a great idea. I would let you know if I made any progress.

...and had you stuck your query into a search engine like google, you
would have found this:

https://code.google.com/p/dumpana/

axel.